An Experimental Investigation of Perturbations on Vortex Breakdown

Document Sample
An Experimental Investigation of Perturbations on Vortex Breakdown Powered By Docstoc
					16th Australasian Fluid Mechanics Conference
Crown Plaza, Gold Coast, Australia
2-7 December 2007

          An Experimental Investigation of Perturbations on Vortex Breakdown over
                                      Delta Wings

                                                    S. Srigrarom and M. Ridzwan

              School of Mechanical and Aerospace Engineering, Nanyang Technological University (NTU)
                        50 Nanyang Avenue, 639798, Singapore. E-mail:

Abstract                                                                (LEV) that are formed by the roll-up of vortex sheets. The flow
An experimental investigation on vortex breakdown on delta              separates from the leading edge to form a curved free shear layer
wings at high angles of attack is presented. As suggested by            above the suction side of the wing, creating a core with large
previous works, perturbations are used to change the platform of        axial velocity components. These large components are due to
the delta wings to reduce the detrimental effects of vortex             very low pressures in the vortex core, which generate additional
breakdown brought about by the Self-Induction Theory. Different         suction and lift force on the delta wings.
patterns of ‘round’ perturbations are tested to obtain the
favourable lift and drag characteristics for each wing. With the        However, at a sufficiently high angle of attack, these leading-
best pattern identified later, optimization of the shape of             edge vortices experience a sudden disorganization or state of
perturbation is explored to further improve the results. ‘Teardrop’     ‘burst’, widely known as ‘vortex breakdown’ often seen on
and ‘diamond’ perturbations are introduced as basis of                  aircrafts as vapour trails. This degradable phenomenon can
comparison. Force measurements were conducted over a range of           severely limit or even eliminate the lift gains achieved by the
α = 0 to 40° to justify the concept of surface shaping and evaluate     platform. The vortex breakdown phenomenon can be generally
its effectiveness. Dye flow visualization were used to obtain           characterized as a rapid deceleration of both the axial and swirl
sectional views of the leading-edge vortices as they break down         components of the mean velocity and the extensive expansion of
for a series of delta wings having sweep angles of 60°, 65° and         the vortex core. During the breakdown process, the mean axial
70°. The wings are tested constantly at a low speed of U∞ = 0.05        velocity components rapidly decelerate into a stagnation point
m/s in a water tunnel facility.                                         and/or become negative on the vortex axis. This stagnation
                                                                        point, identified as the ‘vortex breakdown location’, is unsteady
A combination of side and plan views provides information on            and naturally fluctuates about some mean position, in the
the three-dimensional nature of the vortex structure before,            streamwise direction.
during and after breakdown. Details of the flow at α =15° for
every wing are identified in still photographs while the dynamic        Flow fields around delta wings at moderate to high angles of
characteristics of the breakdown process are examined from              attack have been examined for the past 50 years. Flow
recorded high-speed movies. The force measurement supported             visualization techniques of Lambourne and Bryer [1] have
by the flow visualization shows that certain combinations of            revealed two types of vortex breakdown, i.e. the so-called
perturbations indeed provide encouraging results. For wings with        ‘bubble’ type and the ‘spiral’ type of vortex breakdown.
perturbations, generally, the vortex structure transforms from a        Subsequent research has shown that the spiral type is more
linear structure to a wavy or “kink” structure which effectively        common for delta wings, although occasionally, the ‘bubble’
delay or even suppress vortex breakdown. Various results have           type of vortex breakdown switches to the ‘spiral’ type from time
shown an increase of approximately 10% in lift characteristics          to time in experiments. Thereafter, Srigrarom and Kurosaka [3]
and delay of stall angle for certain scenarios. The best results        have decided to focus on this ‘spiral’ form of the breakdown,
have been for the 60° wing where the ‘teardrop’ bulge in a mild         instead of the axisymmetric ‘bubble’ breakdown. They have
perturbation pattern managed to improve lift characteristics by         formulated a theory called the Self-Induction Theory, relating to
about 15% over the whole range of angle of attack for the tests.        the process of vortex breakdown over delta wings and suggested
Results for the 65° wing and 70° wing are generally positive with       surface shaping to suppress vortex breakdown on delta wings.
the ‘teardrop’ perturbation again providing the best results,           Surface shaping by the introduction of perturbations in the form
however with existence of discrepancies over certain angles of          of ‘bulges’ gives the ability to delay or even eliminate vortex
attack.                                                                 breakdown downstream of the wing. This is due to the departure
                                                                        of the originally linear structure of the vortex core to the wavier
                                                                        or ‘kink’ structure over these perturbations. Hence, eventually
Introduction                                                            the leading edge vortex core diffuses away instead of break
The Highly swept wings, commonly called delta wings due to              down, limiting the detrimental effects of vortex breakdown over
their triangular platform, are used in a variety of aerospace           an unperturbed delta wing platform.
vehicles. Their usage varies from modern combat aircraft to
miniscule unmanned aerial vehicles (UAV). At high angles of             Objective
attack, delta wings are capable of generating higher lift compared
to conventional rectangular wings, with better aircraft stability       The objectives for this experimental investigation are to explore
and acute control characteristics, giving rise to higher                mechanism-based control methodology to achieve radical
manoeuvrability.                                                        aerodynamic gains in delta wing applications such as Unmanned
                                                                        Aerial Vehicle (UAV) by suppressing the performance-limiting
                                                                        phenomenon of vortex breakdown. In addition to that, the author
The flow over a delta wing at high angles of attack, α, is              will measure & analyse force/torque measurements for different
dominated by two large, counter-rotating leading-edge vortices          suppression (perturbation) patterns and shapes of perturbation

using flow visualization to reduce or delay vortex breakdown.
Eventually interpretation of data into lift and drag coefficients            Wing model description
are conducted and analysed for each wing. Recommendation of                  Perhaps the most important component of this study is the delta
the optimum pattern or shape of perturbation (bulges) will be                wing models itself. The clear plastic delta wing models used in
discussed for each wing.                                                     this study have swept angles of 60, 65 and 70°, with sharp
                                                                             leading edges and bevelled at 60°. The chord length, c is 30cm
Scope                                                                        and 1-centimetre intervals are marked along the root centerline.
                                                                             A schematic diagram of the delta wing used in the study is shown
The experimental investigations are conducted at a constant
                                                                             in Figure 2 below.
freestream speed of U = 0.05m/s in a water tank facility in NTU.
The models used in this study are delta wings with sweep
angles, Λ= 60, 65 & 70°. The force measurements are only
limited to the angle attack range of α = 0 to 40°, with increments
of 5° in between each reading. The force measurements would
be taken in 3 orthogonal axes, i.e. x, y & z directions. The focus
of this study is 4 suppression patterns which have been
identified initially as investigation subjects, resulting in a total
of 125 sets of readings for all wings (including readings of base
wing, i.e. no perturbations). With optimization of the best
pattern gathered from experimental results, 2 more perturbation
(bulges) shapes are evaluated as basis of comparisons, i.e.
“teardrop” and “diamond” perturbations, in addition to the
original “round” perturbation, resulting in an additional 54 sets                  Figure 2. Schematic diagram of delta wing model used [21]
of readings. Dye flow visualization is conducted via a gear
pump system with a singular dye probe positioned very near to                The wings used are 0.7cm thick and an L-shaped rod is mounted
the apex of each delta wing, with a constant angle of attack of α            on the leeward side of the delta wing model at mid wingspan so
= 15° for all wings. Video recordings and still photographs                  that it pivots equally through the pitching range. Only the leading
would be evaluated.                                                          edge vortex closest to the tester is used in this study, i.e. the
                                                                             windward side.
Experimental Setup
Water Tunnel                                                                 Perturbation patterns
The force measurements and flow visualization tests are                      In this study, the author used 5 perturbation patterns
conducted in a water tunnel facility at Nanyang Technological                incorporating the similar small and big round bulges. The 5
University (NTU) in Singapore. The cross section of the water                patterns include:
tunnel is 0.45 x 0.6 meter square and the length of the test section         1. base wing pattern (i.e. without any bulge)
is 1 meter, allowing unrestricted viewing of the model tested. The           2. mild perturbation pattern, which only involves the small
tunnel is recirculating, powered by an axial pump at the end and                   round bulge alone at 4cm chord length, equidistant from the
the maximum flow is about 0.17m/s.                                                 root centreline
                                                                             3. strong perturbation pattern with the two round bulges
General layout of Test Section                                                     located along the predicted trajectory of the vortex core at 4
The setup of the model, together with its electrical instruments,                  cm and 6cm chord length, with the small bulge nearer to the
the turntable and load cell are arranged in a manner shown below                   apex
in Figure 1. The dye pump is located opposite of the electrical              4. strong perturbation pattern A, with the small round bulge
instruments to prevent obstruction due to its long hoses                           located near the root centreline and the big round bulge
containing the dye. An illuminated board is attached at the rear                   located near the leading edge at 4cm and 6cm respectively
side of the test section to provide clearer views of the tests               5. strong perturbation pattern B, with the small round bulge
conducted. A delta wing with chord length of 30cm (with its lee                    located near the leading edge and the big round bulge
side facing the tester) is vertically attached by an L-shaped rod                  located near the root centreline at 4cm and 6cm respectively
which is linked to the load cell and turntable. The vertical                 The bulges are always attached in pairs using temporary adhesive
distance of 0.37m is chosen, so as to position the delta wing in             clay for easy removal and the schematic diagram of pattern A &
the middle of the test section to prevent effects of boundary layer          pattern B is illustrated below in Figure 3. The other patterns are
separation. The horizontal distance of 0.15m from the delta wing             illustrated later in Figure 7, with the omission of the base wing.
to the load cell is also deliberately chosen to prevent the vortices
generated to affect the electronically sensitive force

                                                                                  Figure 3. Schematic diagram of Pattern A (left) & Pattern B (right)

                                                                             Bulge nomenclature
                                                                             Another vital component of the experimental setup is the bulges
                                                                             that are attached on the lee side of the delta wing. The small and
                                                                             round bulges are used in the first stage of the tests under the 4
                                                                             perturbation patterns for all delta wings. With a smooth surface,
                                                                             they are designed to minimize the drag effects induced by its
                                                                             presence itself and also to perturb the vortex core in an orderly
                                                                             manner. A schematic diagram showing the dimensions real of the
                                                                             round bulges used are presented in Figure 4.
                 Figure 1.Schematic diagram of test section.

                                                                            This is due to the orientation of the delta wing model during
                                                                            pitching and the pre-determined directions of x, y & z of the load
                                                                            cell in relation to the test section as shown below in Figure 6. The
                                                                            lift and drag forces are always orthogonal in aerodynamics and
                                                                            they are derived from similar orthogonal Fx and Fy forces
                                                                            measured by the load cell.

                Figure 4. Schematic diagram of round bulges used

In the later stages of optimization of bulge geometry, the author
has identified two more bulges with a “teardrop” and “diamond”
shapes with similar dimensions to be used as perturbations.
These two shapes are recommended over other geometrical
shapes such as ovals or ellipse due to its sharp front end, with
the potential to perturb the vortex core more effectively and
cleanly. The dimensions are demonstrated in Figure 5 below.

                                                                            Figure 6. Schematic diagram of test section axes direction (top) & lift and
                                                                            drag forces on delta wing model (bottom)

Figure 5. Schematic diagram of teardrop bulge (top), diamond bulge          The lift and drag forces derived are relatively useful in gauging
(bottom)                                                                    of the forces acting on the model. However a very common mode
                                                                            of comparison for the performance in aerodynamics are the well-
Test conditions
                                                                            defined lift and drag coefficients, CL and CD respectively and
With favourable results obtained at low speeds as observed by
                                                                            these are used for evaluation in the next section.
Lewpiriyawong [2] in his previous studies, the tests were
conducted at a low freestream speed of U=0.05m/s. This
                                                                            Results and Discussion
corresponds to a frequency of about 14.47Hz for the axial pump
of the water tunnel. A conversion reference between the water               60° delta wing
tunnel’s frequency and the actual velocity is included in the               The 60° delta wing generally produces the highest amount of lift
appendix section. The Reynolds number based on chord length                 compared to the other two delta wing models, due to its high
is of 15,000. It is important to note that the water tunnel did not         circulation. The mild perturbation pattern has shown the most
move water during the testing, i.e. the water level does not                prominent results for this particular wing, although the stall
fluctuate about the marked level of the test section. Hence                 angle does not vary that much from the base wing results, as
before every force measurement is started, the water is allowed             shown in Figure 8. With the optimization of the bulge geometry,
to settle for a few minutes. The dye probe is located very near to          the 60° delta wing significantly favours the “teardrop” bulge,
the apex of the delta wing to allow diffusion of the dye on the             providing around 10% increase in CL throughout the range of
vortex core which forms up very near at the apex leading edge.              angle of attack, as shown in Figure 9. During flow visualization,
The temperature of the water in the tunnel is also kept constant.           the “teardrop” bulge too manages to turn the initial straight
                                                                            vortex core to a very undulating one. This helps to recover the
Data Conversion                                                             lift loss during pitching and also suppresses vortex breakdown
The data recorded in excel format are mere numbers with units               as the linearly inductive vortices are deviated in such a
in Newtons. It is almost impossible to evaluate or analyse the              manner.(Figure 10) Hence, the results of the study for the 60°
readings from force readings only. It is useful only when                   delta wing clearly favour the “teardrop” bulge as its mechanism
converted to lift and drag coefficients which are primary basis of          for perturbation.
comparisons in aerodynamics. Only through this method, the
author is able to judge the reliability of the setup and                    65° delta wing
instruments and also the extent of success of the bulges to                 The 65° delta wing also produces similar results, with the mild
perturb the vortex core by comparing with published works. The              perturbation pattern producing the highest CL value for its case
main focus of the author is the Fx, Fy and Fz readings, which are           after α = 17°, during its tests (Figure 11). However, more
the load forces in the x, y & z directions. The lift and drag               importantly, the mild perturbation for this particular wing,
components of the delta wing are related with these 3 forces                manages to delay the stall angle by a further 10°. This shows
mentioned together with the angle of attack, α, by the following            that the mild perturbation is very effective for the 65° delta
equations (unit in Newtons):                                                wing. The results from the optimization of geometry stages also
                                                                            indicate remarkable CL values for the “teardrop” bulge,
               Lift , L = F y cos α − Fx sin α                              producing increase in CL of more than 10% and also a delay of
                                                                            stall angle of more than 10° (Figure 12). Results from the flow
              Drag , D = F y sin α + F x cos α                              visualization tests also reinforce the effectiveness of the
                                                                            “teardrop” bulge, providing the most undulating wavy vortex

core. (Figure 13) Hence a mild perturbation, utilizing the
“teardrop” bulge is most suitable for the 65° delta wing.

70° delta wing
Force measurement results (Figure 14) for the 70° delta wing
shows that the stagger patterns A and B are particularly
ineffective in perturbation. Instead, the two patterns actually
decrease the amount of lift created, resulting in very low lift
coefficient values. Similarly, to the other two wings, the mild
perturbation again provides good results, particularly in the                         Mild perturbation                         Strong perturbation (along
range α = 20°- 30°, providing increase of more than 10% in CL                                                                   trajectory)
values. The flow visualization results from the 70° delta wing
rather have been rather peculiar, if compared to the other two
wings. This is probably due to its small apex area, resulting in
lower circulation. The “teardrop” bulges located on the 70° delta
wing have only managed to perturb the vortex core slightly,
resulting in a small diameter of the ‘kink’ structure (Figure 15);
it usually produces for the other wings. On the contrary, the
“diamond” bulge produces more undulation in the wavy vortex
core and this is reflected in its favourable values during
optimization of bulge geometry tests (Figure 16). Hence a mild
perturbation, in this time round, using the “diamond” bulges is                   Strong perturbation pattern A               Strong perturbation pattern B
more efficient and effective.
                                                                                        Figure 7. Actual perturbation patterns of round bulges
Conclusion & Recommendations
With the exploration of the different perturbation patterns and                                      60deg wing:Effects of perturbation on CL
also the three types of geometry used, the objectives of the study          1.2
are clearly achieved. However, this is not an exhaustive study,
as the field in flow analysis for vortex breakdown is boundless.
Therefore, further improvements can still be recommended and                 1
evaluated to achieve a more comprehensive goal. Firstly, the
size of the bulges has not been varied as a free variable. In other
words, a smaller or bigger “teardrop” bulge may or may not                  0.8
produce good results for the three wings. Hence, this is a
particular area that can be worked on. Finally, the location of the
bulges is also not varied along the root chord line. The bulges             0.6
were fixed either in the 4cm or 6cm chord mark. A further study

can look into this particular matter to further improve the
performance of the perturbation methodology, in line with the               0.4
Self-Induction mechanism theory. Perhaps locating the bulges
                                                                                                                Mild perturbation
nearer or further away from the apex have different reactions to                                                Base Wing
the linearly inductive vortices. Even more variables can be                 0.2
                                                                                                                Strong perturbation(pattern B)
recommended in the pursuit of understanding better the roles of                                                 Strong perturbation(along trajectory)
perturbation on the delta wing.                                                                                 Strong perturbation(pattern A)
                                                                                  0       5     10        15         20             25       30         35   40
References                                                                                                      angle of attack

[1] Lambourne, N.C. and Bryer, D.W., “The Bursting of                         Figure 8. Lift coefficient of 60° wing for 5 perturbation patterns
    Leading-edge vortices-Some Observations and Discussion
    of the Phenomenon”, ARC R&M 3282, April 1961
[2] Lewpiriyawong, N. and Srigrarom, S., “Modification of
    Delta wings to control vortex breakdown”, Proceeding of
    the 12th International Symposium on Flow visualization,
    Gottingen, Germany, 2006
[3] Srigrarom, S. and Kurosaka, M., “Surface shaping to
    suppress vortex breakdown on Delta wings”, AIAA Journal,
    Vol.38, No. 1, 2000

                   60deg wing:Effects of perturbation on CL                                                            65deg wing: Effects of perturbation on CL

1.2                                                                                         0.8




0.6                                                                                         0.4

0.4                                                                                                                                      Base Wing
                                            Mild perturbation(teardrop bulge)               0.2                                          Milf perturbation
                                            Mild perturbation(diamond bulge)                                                             Strong perturbation(along trajectory)
0.2                                                                                                                                      Strong perturbation(pattern B)
                                            Mild perturbation(round bulge)
                                                                                                                                         Strong perturbation(pattern A)
                                            Base wing
                                                                                                    0       5          10           15         20        25            30        35   40
      0    5      10        15       20          25        30          35       40
                                   angle of attack                                                                                       angle of attack

  Figure 9. Lift coefficient of 60° wing for 3 types of bulge geometry                     Figure 11. Lift coefficient of 65° wing for 5 perturbation patterns

                                                                                                                            65deg wing: Effects of perturbation on CL
                                                                                                        Mild perturbation(teardrop bulge)
                                                                                                        Mild perturbation(diamond bulge)
                                                                                           1.2          Mild perturbation(round bulge)
                                                                                                        Base wing






                                                                                                    0       5          10           15          20          25          30       35   40
                                                                                                                                         angle of attack

                                                                                                Figure 12. Lift coefficient of 65° wing for 3 types of bulge geometry

          Figure 10. Visualization result of 3 different
          bulge geometry on 60° wing

                      Figure 13. Visualization result of 3 different bulge                            Figure 15. Visualization result of 3 different bulge geometry on
                      geometry on 65° wing                                                           70° wing

                                    70deg wing: Effects on perturbation on CL                                                70deg wing: Effects on perturbations on CL
               Base wing                                                                                       Mild perturbation(diamond bulge)
               Mild Perturbation                                                                               Mild perturbation(teardrop bulge)
               Strong perturbation(along trajectory)
               Strong perturbation(Pattern A)
                                                                                                     1.2       Base wing
     1         Strong perturbation(Pattern B)                                                                  Mild perturbation(round bulge)








         0        5            10           15          20          25    30    35   40
                                                                                                           0    5          10          15            20           25   30   35   40
                                                  angle of attack                                                                               angle of attack

         Figure 14. Lift coefficient of 70° wing for 5 perturbation patterns                     Figure 16. Lift coefficient of 70° wing for 3 types of bulge geometry


Shared By: